Machsupport Forum
Third party software and hardware support forums. => SmoothStepper USB => Topic started by: phongshader on October 24, 2015, 11:00:51 AM
-
I've searched for this info but have found next to nothing. The documentation of the CNC4PC c25 board says the way to wire up a probe for auto zero is to run an input wire to the bob on pins 10-13, 15 and run a 4.7k ohm res. from 5vdc pin to that input pin, and to run a wire to a bob grnd pin. What I've done is run an input wire to port1, pin15, bridged the 5vdc pin (between pins 12-13) and pin15 with a 4.7k ohm res. and run a wire to the bob grnd pin(between pin 15-16). In Mach3 in the pins and ports dialogue I've enabled the probe, set the port to 1 and the pin to 15 and left the active low and emulated to the default, which I think means it's set to active high? When I short the probe wires there is no indication that the probe is active....so what did I do wrong? I'm running the 2010 screen set, Mach3 3.043.062, and ESS v10r2d1d plugin.
Take 2:
When in the ESS Data Monitoring utility there is no response on pin 15 (in the data monitoring utility) when I short the lead from pin 15 to board GND. The homing sensors are working as expected, pins 11-13 are checked in the data monitoring utility when the sensors are triggered and unchecked when not triggered.When I put a voltmeter from GND to pin 15 (bridged to 5v through a 4.7k ohm resistor) it reads .49v and when I short the lead from pin 15 to GND the voltmeter drops to 0v...this is the expected result, right? The only thing wrong, in my uneducated guess, is that I'm not getting 5v untriggered, is this correct? This being the case, activation would not show up in the data monitoring utility because it is looking for a 5v input, correct? I'm not an electronics expert or even a newbie so I'm guessing here, is the 4.7k ohm resistor from the 5v pin( between pin 13 and 14) and pin 15 the correct resistance? How would or do I get 5v to pin 15? Or do I need to? to add to this I would need to change the probe pin configuration to active low, correct?
Thanks,
Justin
-
Here is a schematic and photo of wiring implementation.
-
I changed the 4.7k ohm resistor to a 2.2k resistor and now the probe works...sort of.
I'm using the simple auto tool zero in the Mac3 2010 screenset.
- When not grounding the tool/spindle and just touching bob grnd to touch plate it works, but stutters a little bit on the first contact but finishes the routine and reads "Z axis is now zeroed"
- When grounding the tool/spindle with the bob grnd the routine runs and when the tool touches the touchplate it hesitates for a microsecond, like it did in the above scenario, but then continues down, even though it has touched the plate and the probe led is lit, the screen still says "probing for zero".
I have to hit the E-stop to stop the probing routine.
There is continuity between the bob grnd and the spindle/machine at about 9M ohm. I don't know if this make a difference but I've heard that sometimes the spindle/tool needs to be isolated from the bob grnd. I've also heard that noise is usually the culprit in these situations. If it is noise wouldn't the noise trigger a false reading, stop the routine early and NOT ignore a real reading of a touch? Looking at the schematic in the previous post how would I add some noise suppression? Anyhow I'm out of my depth here and any input or suggestions would be greatly appreciated.
Thanks,
Justin
-
What version of Mach3 are you running?
The recommended version with the ESS is ...062 and not 066
I suspect that your problem (in part) is associated with the input schematic for that board.... If you look at the at the attached screenshot extract...
You basically have a voltage divider being formed .... so when you apply 5V to pin 15 via a 4.7k resistor.... the voltage seen at pin 15 is 2.5V
<<<Ohms law >>>
by reducing the resistor to 2.2K.... you've increased the voltage to ~ 2/3rds of 5v, which is 3.4V
The datasheet for the Schmitt Trigger is here .... have a look at the switching voltages on page 5/21:
http://www.nxp.com/documents/data_sheet/74HC_HCT14.pdf
I would suggest using a 1kohm resistor.... presuming you bought a pack of 2.2k ohm resistors.... I'd join tow together in parrallel (side by side) ..... NOT SERIES!!....
As that would make the high side voltage 4.05V which is well above the threshold voltages...
(really they should have used a jumper so you could have removed the 4.7K resistor).
You can set your debounce quite short as the inputs have capacitors (especially for your "probe" input on the ESS config screen) .... hence when the probling function starts the capacitor will be charged to ~ 4.05 volts (as per my calc above).... and when the probe is grounded the capacitor will discharge in about 470 micro seconds
-
.... note you must ground your machine when you are doing the probing function...
Rob
-
Oh dear...... Ignore my above post....... I've made an error (over the grounding comment!!!!) [[[distractions in the background whilst trying to type and multitask!!]]]
-
Thanks for the advice, I added a 2nd 2.2k res in series and now the touch plate is reading @ 4v however I still have the following issues:
- When not grounding the tool/spindle and just touching bob grnd to touch plate it works, but stutters a little bit on the first contact but finishes the routine and reads "Z axis is now zeroed"
- When grounding the tool/spindle with the bob grnd the routine runs and when the tool touches the touchplate it hesitates for a microsecond, like it did in the above scenario, but then continues down, even though it has touched the plate and the probe led is lit, the screen still says "probing for zero".
I have to hit the E-stop to stop the probing routine.
-
How are you running the probe function.. Which mcode are you using or script
Have you got your machine setup correctly where
Z+ = up
Z- = down
Rob
-
Yes z+ up z- down, I've had this working before, don't why it stopped working.
-
Have you got homing / limit switches .
What pins hae you got setup for the probe and homing functions?
(I am trying to remember what I changed 4 months ago when I had the same problem)
Rob
-
Try changing the z axis settings on the homing and softlimits settings page under configuration...
There is a checkbox for negative home, can't remember if toggling that solved my problem (I had two at the time the z was crashing int o the top of the carrige but that was the safe home setting...)
-
yep, I've got homing switches that work, x Home pin 11, y Home pin 12 , z Home pin 13, probe pin 15, all on port 1. Homing is active high and probe is active low.
-
Try changing the z axis settings on the homing and softlimits settings page under configuration...
There is a checkbox for negative home, can't remember if toggling that solved my problem (I had two at the time the z was crashing int o the top of the carrige but that was the safe home setting...)
I'll try that but I don't think that's the issue. the auto zero macro works, more or less, when the C25 grnd is NOT touching the tool/spindle and I touch it to the touch plate, that is I start the macro, z goes down towards the table, -z, until I touch the bob grnd to the touch plate, z stops then it does it's double touch routine then retracts to 1.5" above what it thinks is zero. Now when the bob is grounded to the tool/spindle and I run the macro when the tool hit the touch plate it keeps on going even when the probe led is lit and I have to hit the e stop to keep from crashing into the table.
-
what version of Mach3 are you using?
(other than that I'm out of ideas).
-
what is the code you are using if it`s not coded for you screen it wont work to well put it in code tag's
-
what version of Mach3 are you using?
(other than that I'm out of ideas).
Version R3.043.062
what is the code you are using if it`s not coded for you screen it wont work to well put it in code tag's
I'm running the 2010 screenset.
Like I said before I'm totally new to all this, especially the electronics part. A little back story here. I bought a Chinese machine with an AKZ250 motion controller card, set up the auto zero probe with an external power supply (9 volt battery) and it worked, and then one day it didn't. I swapped out the AKZ250 card for an ESS/C25 combo and the auto zero still didn't work, played around with it, re calibrated the motor tuning and it started working, just not consistently. I moved the touch plate grnd lead from the front of the table to up onto the spindle mounting plate and the auto zero stopped working. Now if I were reading this I would think that this would be a noise issue because now one of the probe leads is running past another stepper motor and a spindle however I believe there is pretty good noise suppression on the C25 board and I'm running 5v through 2 2.2k res in series to the touch plate. What I am thinking now is that there is a grnd loop issue because the auto zero macro runs as it should when the C25 probe leads/touch plate is isolated from the table, spindle, tool, and doesn't when the grnd lead from the C25 is touching the table, spindle, tool. Does this make sense? If this is indeed a grounding issue how would I go about fixing it, or even testing it beyond the test I've done?
't
-
.... Can I make a suggestion.... Why not use a crocodile clip onto the tool when you want to use the tool offset / probe function
Because at present you are relying upon conduction through the spindle bearings and spindleshaft to the tip of the tool.
Then put the 5v (pin 13 or 15 input..forget which one you are using) to your insulated calibrated height block.... (You could use a piece of blank PCB (printed circuit board)....
Try that...
-
Why not use a crocodile clip onto the tool when you want to use the tool offset / probe function
I am using a crocodile clip on the tool or when the tool is to big for the clip I clip it to a bolt head on the spindle mount, but for the testing I have been clipping to the tool. To add a little more to my grounding theory if I touch the ground to the tool in the middle of the zeroing macro I can feel the machine react to the contact, not much, but there is something happening and this is long before the tool reaches the touch plate.
Then put the 5v (pin 13 or 15 input..forget which one you are using) to your insulated calibrated height block.... (You could use a piece of blank PCB (printed circuit board)....
you just described my setup as it stands now, although my circuit board is loose so I can move it around to the work piece.
-
so you have a clip that goes to the tool what is earth (common) and the signal wire going to the touch plate.
do you have your earth wires going to a common star point earth is -common or just common.
also what is the voltage between the clip and the plate if it is below 2.8vdc that might be the problem.
if you just touch the plate to the spindle noise does anything happen or the frame of the machine, if you touch the plate to the spindle and it works you don't need the clip if it does not do anything you need the clip.
if it works when you touch the plate to the machine frame good if it is still just as bad, check all you wiring to make sure you only have single ended shield wire`s no broken wires just a good look over.
gerry's code is fine it will be nothing to do with that.
a ruff way to check to see if it is a noise (RF) problem put a fm radio on the machine (have it turned off) tune to a radio station turn the machine on and if it goes stac you have a noise problem if the radio is fine it`s to small for the radio to pick up so all good.
-
I suspect it's a ground loop problem.... You may need to try to put the ESS on its own 5v power supply so as to separate both sides of the breakout board optoisolators.
Have you got a separate 5v power supply you could try?
Rob
-
the ESS should be on it's own 5vdc line with a common (earth) it the best way to do it, a lot of noise problems I was having went away after I rewired the whole machine to a common and it`s own 5vdc line for the ESS.
and moved the high and low common and power wires so there was no side by side wires, a cross over at one spot what is fine and the high and low common have there own star point side by side what is also fine.
-
so you have a clip that goes to the tool what is earth (common) and the signal wire going to the touch plate.
ground is the ground pin on port1 between pin 15-16 and signal wire is pin 15 on the C25 board
also what is the voltage between the clip and the plate if it is below 2.8vdc that might be the problem.
4v
Have you got a separate 5v power supply you could try
I do have a separate 5v power supply, you can see it in the attachment
the ESS should be on it's own 5vdc line with a common (earth) it the best way to do it, a lot of noise problems I was having went away after I rewired the whole machine to a common and it`s own 5vdc line for the ESS.
and moved the high and low common and power wires so there was no side by side wires, a cross over at one spot what is fine and the high and low common have there own star point side by side what is also fine.
The power supply I installed for the ESS is grounded to the chassis, common to the earth ground of the 220v from the house. What I'm unclear about is there are now 2 grounds, one from the C25 board and one to the chassis/earth from power in from the house. The power supply I installed has line, neutral, and ground, the ground on the ESS is derived from 5vdc- from the 5v power supply I installed, where s the common ground for the vdc? The chassis is grounded VAC, the ESS/C25 is VDC. I'm probably not understanding how the grounding works but hopefully some one can straighten me out.
-
you have not got your commons connected quite right -common is earth, ground you can have a +common (what is Power) and a -common look it up use google you need to learn this it is important that you learn how to do it.
-
You have a major ground loop problem....
Remove your power supplies 0v outputs from the chassis earth for the control box
So just feed the ESS and bob with +5v and 0v
And just feed your stepper drives with +v and 0v
Then try your probe routine.
Rob
-
If your machine is earthed .... By clamping pin 13 or 15 of your bob to your spindle you are essentially making the bob power supply +5v = +1v (same potential as the control box chassis) and the 0v is -4v ( below earth voltage)...
You have two isolation transformers there and you do not need to connect the secondary side 0v lines to earth, control box chassis earth or the machine earth (all earth's should be the same voltage)
It sounds like if you machine was earthed... When you claimed the pin 13/15 from the crocodile clip onto your spindle... You were also trying to apply +4v (because of the voltage divider I noted above) onto the same 0v line.
Rob
Rob
-
bang on Rob, (the common to the earth ground of the 220v) this ground should be sunk to earth as not connected to the machine but being connected to the earth stack as it will be, you only earth the main power supply wire if you are told to by a electrician you can wrap a shield wire around it what would go to the high voltage common.
also what I said above was checked by an electrician
-
Earthing is very controversial (and emotive) .... Many countries have different ways of doing it and providing an earth (each think they are right)
... Long and the short ...
If you have an ac supply from the street (electric supply company transformer) ... It needs an earth ... As the transformer in the street will be earthed ... So that if you measure between any of the incoming wires none of them will exceed the supply voltage relative to earth...
Does everything need to be earthed... That depends on that voltage is going into the object relative to earth, and also how the supply is wired to detect current flow to earth (should a problem occur and something becomes a higher voltage relative to earth) .... In the UK the use of residual current disconnection devices has become very common in some scenarios so that current flow between the two current carrying wires is monitored and should a difference occur above a prescribed amount for a prescribed time the supply is disconnected.
It is one reason why I choose not to earth my machine but my control box is earthed. (Another is I've had more 230v shocks in my life than I care to admit...and I'm still here... (Some were not my error I must point out the most interesting of which was when I was on a ladder and touched the top of a metal stud wall to steady myself with one hand and used the other to try to lift a metal ceiling tile... The funny thing was my brain had to figure out what the sensation was ... Only for me to confirm it by checking it for a second time ...
"Electrician never had an electric shock has never done any work"... It gives you a healthy respect for moving slower and thinking more when you are working on 415v or above when your room for error diminishes...)
If you are vetting an electrician ask to have a look in his toolbox ... If the tools are old but show little or occasional old flash marks ... He's relatively safe and experianced... If the tools are new ... He's inexperienced or covering something up ... And if they show recent flash / burn marks ... He's still learning...
Rob
-
sounds about right
-
So this is a lot to understand but what I'm getting as the take away is that I need to eliminate the 220v earth/ground. The 5v power supply I installed does not want to share it's ground with 220v earth. I've attached a schematic of how the machine is wired at the moment. I added, I guess foolishly, the 220v earth, that was not originally installed when I got the machine. Any how Could you please look at the schematic and point out the flaws in grounding. You talk about 0v I'm unclear as to what is 0v in the schematic the 5v power supply has a line, neutral and ground input and a v+ and v- out where is the 0v in the schematic?
Thanks again for helping me out!!
-
What you have drawn is correct... Except don't remove that earth...
What I was talking about (the 0v) is what you have drawn as v--... Alot of people connect them together and then connect them to the chassis casing via the earth lug
I would suggest you go through your wiring and double check you have wired it as you have drawn it
Sorry need to go to day job to pay the bills
Rob
-
Your e-stop is not in the right place as shown it is more of a kill switch and your ESS and Mach will not be happy
Rob
-
I most definitely did not ground the -v/0v of the power supply to the chassis 220vac earth, or at least not intentionally.
I measured the voltage between the C25 ground and the grounded spindle and it is 0v.
I'll check in the morning... must sleep now.
-
it should be 0v sitting there, as rob say your E stop is a kill your stepper driver switch you just need it for the 5vdc power supply no power to ESS or C25 no motion.
killing the power to the stepper motor when they are under load can cause a power surge back to the stepper driver and magic smoke will come.
now just for fun touch the wire attached to the plate to the clip there may be some stuff on the plate giving a bad contact. the voltage is within the amount that it needs to work.
also just check for continuity between plate and clip with no power to the machine.
one other thing have you changed the pin from active high to low or vices versa the red x to green tick
-
I went through and checked continuity:
0 continuity between chassis grnd and C25 grnd pin
0 continuity between chassis grnd and C25 +5v pin
20.5M resistance between chassis grnd and table/spindle/tool
0 resistance between C25 pin 15 (probe signal) and touch plate
0 resistance between C25 grnd and clip
0 resistance between C25 pin 15 and C25 grnd when clip is on touch plate
Measured voltages:
4v between C25 pin 15 and grnd when probe circuit open (clip not touching plate)
0v between C25 pin 15 and grnd when probe circuit closed (clip touching plate)
0v between between chassis grnd and C25 grnd
0v between between chassis grnd and and C25 pin 15
0v between between chassis grnd and C25 +5v pin
I'm going to re wire the Estop as soon as I post this. Thanks for the heads up blockhead and Rob
-
rob the C25 manual have a look http://cnc4pc.com/Tech_Docs/C25R3_1_USER%20MANUAL.pdf
by looking at the C25 manual, you can have switches connected using the 5v of the board to a pin what is simple, the prope touch plate needs to be grounded to the C25 board it has the ground symbol like you said you have it connected to the ground of the board
-
Why not use a crocodile clip onto the tool when you want to use the tool offset / probe function
I am using a crocodile clip on the tool or when the tool is to big for the clip I clip it to a bolt head on the spindle mount, but for the testing I have been clipping to the tool. To add a little more to my grounding theory if I touch the ground to the tool in the middle of the zeroing macro I can feel the machine react to the contact, not much, but there is something happening and this is long before the tool reaches the touch plate.
Then put the 5v (pin 13 or 15 input..forget which one you are using) to your insulated calibrated height block.... (You could use a piece of blank PCB (printed circuit board)....
you just described my setup as it stands now, although my circuit board is loose so I can move it around to the work piece.
I still think you have a grounding issue with a ground loop somewhere as there should be no reaction if you touch the tool to the ground at any point ^^^^^ (read the quote / statement above from an earlier post).
Yup I've been having a look at the manual...thanks
And I've put a post on the Warp9 website (suggest you register and have a look for yourself).
http://www.warp9td.com/
(Andy should be working for a little while longer yet US time)
Rob
-
This scares me, for the last week I've been chasing down this bug of why the auto tool zero was not working with the help of all of you. Going through all these iterations but ultimately changing nothing on the machine, settings, wiring, or otherwise. Out of frustration today after work I went into the shop to see if I could find that elusive ground loop we all know is there with a voltmeter and ran the auto zero routine to see if there was any voltage between the C25 grnd and the chassis grnd while running the routine...none, so just for kicks I let the machine run through the whole routine and....it worked?!?! I repeated it 5 times and 5 times it worked. How is this possible? Like I said this really scares me, for no rhyme or reason it started working so I guess for no rhyme or reason it could stop working. Ahh well I guess I shouldn't look a gift horse in the mouth. ;D We'll see if it's working tomorrow. Thank you all for your patience and help. I'll probably need it again soon. ::)
-
Don't be scared by any of this treat it as an adventure... You never know what's around the corner... Have a healthy respect for it and safety as the machine has no brain (fingers out of the cutting area etc)... Some days stuff works... Some days not... As Edison said " I have not failed. I have only found 1000 ways not to do something "
Glad to hear its working.
May I ask..
What type of network cable are you using (any markings on it)... Is it shielded? ... Cat6?
Does the network cable go directly from the PC to the ESS... Or have you used a lead with a bulkhead connector?
Rob
-
yer it sounds like you have a RF problem as well. having a high grade ethernet cable helps. if this happens again shut everything down in your shed and only have a fm radio going no lights nothing turn the machine on and if the radio becomes stackey you have a RF problem from the machine turn it on and off a couple of times to see if it will clear it self.
by having only the radio and machine on you can eliminate the machine quickly as the source of the RF. then if it`s not there you just turn one thing on at a time to it comes back.
I have a random bad RF problem with my big machine, when the RF is there the radio goes all stackey and the machine run like a bag of poo, quite often I can just turn the machine off and restart and it goes away.
and if I stand in line with the radio it goes away as well that's how I know it's from the machine.
-
Good method there thanks
Rob
-
Andy at warp9 responded to my post saying that all four mounting holes of the ESS are isolated from ground (I was concerned that you may have used metal risers)... But did state that the Ethernet sheild is grounded... So if you use a cat 6 cable (or like I do a fly lead with a chassis / bulkhead connector on the end of it so I can just connect all my wiring to the outside of my control box)... The ESS will form a ground with the casing or PC and this may create a ground loop
... I say may as these things can be occasional an problematic.
Like rf noise
Rob
-
yer its easier than using a TV, when I can afford to get a scope I will be able to find the problem.
the computer needs isolated from the machine or a ground loop will occur just makes sure it's galvanically isolated cable, high quality ones are single ended
-
Thanks again for the all the help.
when I can afford to get a scope I will be able to find the problem.
I do have a scope that I bought for tracking down a problem I had with a fuel injection/ignition system I built a few years back. How would I use the scope to track down rf problems?
-
As I suspected...I got up this morning to check to see if the auto zero was still working...it is not. So I replaced all the leads from the C25 with shielded wire, ran an earth lead from the chassis to the router table put it all back together and it works again...for the moment. My thinking on the ground from the chassis to the router table is that was the only variable between last night and this morning was when I tested last night the chassis was sitting on the router table (aluminum) so, in theory, the table was grounded to the chassis. If there is a good reason not to do this please let me know.
Thanks again.